Prevalence and Associated Factors of Psychological Distress Among Single Fathers in Japan

Background In Japan, ten percent of single-parent households are led by fathers. Taking care of children as a single father is very stressful and could put a strain on their health. It is very important to prevent and identify psychological distress among fathers for both their own health and to avoid negative impacts on children. This study aims to determine the prevalence of and factors associated with psychological distress among single fathers and understand how it is different from partnered fathers. Methods We used data from the Comprehensive Survey of Living Conditions 2016. Psychological distress, assessed using the K6 scale, was analyzed among 868 single and 43,880 partnered fathers. Logistic regression analysis was performed to assess the risk factors for psychological distress, such as employment type, sleep hours, and smoking and drinking habits. Results Single fathers had a higher proportion (8.5%) of psychological distress compared to partnered fathers (5.0%). A larger percentage of single fathers had a lower educational level and were more likely to be non-regular workers, self-employed, or unemployed than partnered fathers. Among single fathers, the crude and adjusted odds ratio for employment type and sleep hours were significantly associated with psychological distress. Conclusion As single parents who are self-employed or directors are likely to have significantly reduced psychological distress than those with regular jobs, measures are needed to improve the work-family balance for non-self-employed fathers. There is a need to provide greater financial assistance and other social welfare support to single parents to ensure their and their children’s good health.


INTRODUCTION
Single parenting is prevalent all over the world, and the proportion of single parents is increasing with the increase in rates of separation, divorce, and premarital childbearing. 1 Men nowadays are expected to be involved during pregnancy, birth, and childrearing activities, in contrast to previous generations. This trend can also be observed in Japan, where the average size of a household decreased by a third from 1970, to 2.4 individuals in 2010. 2 Additionally, the proportion of single-parent households increased from 5.7% to 8.7% from 1970 to 2010, 2 with the increasing rates of divorce. 3 Ten percent of single-parent households in Japan are led by fathers. 4,5 These households have often low-income, and the average annual income of single fathers is only 80% of the equivalized average income of two-parent households. 6 The physical and mental health of single fathers has not received as much attention as that of single mothers, who are more common. 7 Single fathers have been eligible to receive subsidies for rearing their children since 2010, yet they often still face financial difficulties. 8 Taking care of children as a single father is very stressful and could deteriorate their health.
Studies suggest that single parents are more likely to have poor mental and physical health than partnered parents. 7,9 Studies from Canada, the United Kingdom, and South Korea that compared single and partnered fathers showed that single fathers have poorer self-rated health, 10 more depression, 11 and poorer quality of life. 12 This makes single fathers likely to commit suicide in a country like Japan, where suicide rates have remained high for a long time. 13,14 In comparison to partnered fathers, single fathers have less social support and are more likely to smoke tobacco and drink, which is associated with three-fold higher mortality rates in single fathers compared to partnered fathers and single mothers. 15 This worse health condition is associated with single fathers' lower socio-economic condition. 7 Depression among single mothers and its effect on child development is well known. 16 A growing body of research suggests that paternal depression elevates the prevalence of childhood depression, 17 as it is associated with unwillingness to involve in child-rearing activities. 16 Depressed single fathers have been found to use physical disciple on children more often than non-depressed fathers. 18,19 Thus, it is very crucial to prevent and identify depression among single fathers to avoid negative impacts on both them and their children. Attention needs to be given to such single fathers, as people with mental illness experience many preventable diseases, such as diabetes, coronary heart disease, and metabolic syndrome. 20,21 Family support services were started in Japan in 2009, as per the Child Welfare Act. 22 However, men are often reluctant to access healthcare services. Compared to single mothers, they are more often found to report poor self-rated mental and physical health but are less likely to seek medical help. 23 A study found that 80% of the men refused to seek medical help until encouraged by their partners. 24 Growing evidence in the medical literature has identified unemployment, age greater than 30 years at 6 weeks after childbirth, 25 financial instability, older age, work-life balance, and relationship issues with children as risk factors for paternal depression. 26,27 To date, not many studies focusing on the wellbeing of single fathers in Japan have been published, and a few emerging studies have shown that single fatherhood is related to levels of psychological distress at elevated levels similar to those of single mothers. 28 These studies, however, focused on the postnatal period or did not use nationally representative data. It remains unclear what factors are associated with psychological distress and how it differs in single and partnered fathers. It is vital to understand what factors contribute to the development of psychological distress among single fathers, and the distinction of these factors from partnered fathers is crucial to deliver assistance to single fathers. Hence, this study aims to determine the prevalence of and factors associated with psychological distress among single fathers and understand how it is different from partnered fathers.

Data
In June and July 2016, the 11 th large-scale Comprehensive Survey of Living Conditions (CSLC) was conducted nationwide by the Ministry of Health, Labour and Welfare of Japan, collecting data on household, health, income, and living conditions. 29 In this national cross-sectional survey, data from a self-administered household and health survey was used. The questionnaires were distributed to approximately 710,000 people in 289,470 households from 5,410 randomly selected areas from the 2010 National Census. Members from 224,641 households completed the questionnaire (household-level response rate of 77.6%). We obtained permission from the Ministry of Health, Labour and Welfare of Japan to use the data for the intended purpose of this study. We also obtained approval from the ethics committee at the National Centre for Child Health and Development (No. 2020-299).
For the current study, we merged data from the household and health survey questionnaire. We prepared a dataset with data from 568,426 respondents living in 224,208 households. Figure 1 shows how we extracted the study population selecting one father, single or partnered, from each household using the defined inclusion and exclusion criteria. The final sample size consisted of 868 single and 43,880 partnered fathers aged 18 years and above, who had a child aged below 18 years. Table 1 shows the characteristics of the study participants in detail.

Comprehensive Survey of Living
Conditions n=568,426 Households with fathers n=45,632 Household with fathers n=44,748 Single fathers n=868 Partnered fathers n=43,880 Excluded 884 missing outcome variables Excluded 522,794 participants who were not the father of the youngest child aged below 18 in the household thus, selecting one father from each household

Classification of single and partnered fathers
In this study, fathers were classified as single if their household consisted of fathers, who were living with just children or living with children in a three-generation household and defined as partnered if they were married fathers living in a household of a couple and unmarried child, or a three-generation household with children.

Outcome variable
Psychological distress was assessed using the six-item Kessler Screening Scale for Psychological Distress (K6) scale. 30,31 The K6 scale has been tested for reliability and validity after translating into Japanese. 31 For this study, we obtained a Cronbach's alpha of 0.93 for the K6 scale. This demonstrates an excellent internal consistency and reliability. The K6 scale contains questions starting with "During the last 1 month, how often did you feel" and consists of six related questions about feeling nervous, hopeless, restless, depressed, and worthless. Respondents self-report as 0 ("none of the time") to 4 ("all of the time") with a five-point response, the sum of the scores ranging from 0 to 24. Calibration studies have identified a total score of 0-7, 8-12, and 13-24 to represent low, moderate, and highlikelihood of psychological distress, respectively. 30,32,33 For the current study, a total K6 score of 0-12 was classified as without psychological distress and a score of 13 and above as having psychological distress according to the recommended cut-off points. 32,34,35 Explanatory variables We chose employment type, working hours, outpatient department (OPD) visit except for depression, medical exam during the past year, sleep hours, smoking and drinking habits, type of home, family type, household expenditure per person per month, age, education, child's age, and the number of children as explanatory variables based on findings from previous studies. 26,27,33,36,37 Due to the low number of observations in each original variable category for single fathers, most of them have been recategorized into fewer categories. Further details on the re-categorization of explanatory variables can be obtained from eMaterials 1.

Statistical analysis
The proportions of participants, overall as well as separately for single and partnered fathers, were calculated for all variables of interest. Similarly, the proportions of participants with psychological distress were calculated separately for single and partnered fathers for all variables. Data are presented as percentages. We used fisher's exact test or chi-square test to compare categorical variables and obtained the P-value to examine the difference in the proportion of variable categories for presence and absence of psychological distress. We used logistic regression analysis with listwise deletion for missing observations to estimate the odds ratios (ORs) and 95% confidence intervals (CIs). We initially conducted a bivariate analysis, without any adjustment. Then we applied a logistic regression model for the outcome variable psychological distress and adjusted for education, employment, OPD visit except for depression, medical examination during the past year, sleep hours, drinking and smoking habits, type of home, family type, household expenditure per person per month, age category of youngest child, and number of children. Father's age was not included in the final model, as it was collinear with the child's age as determined using Pearson's correlation coefficient of 0.6. Each analysis was conducted separately for single and partnered fathers. With increasing age as children enter different phases of adolescence, the varying social context surrounding children may have had some contextual effects 37 on the psychological distress among single fathers. The explanatory variables could be differently associated with psychological distress depending on whether the child is in the preschool age (0-5 years old), elementary school age (6-11 years old), or adolescent age (12-17 years old). To understand the situation better, we analyzed the descriptive statistics of single fathers separately by child's age (eTable 1). All analyses were conducted using Stata MP 14.2 (Stata Corp, College Station, TX, USA). Table 1 shows the descriptive statistics for single and partnered fathers. The distribution of most of the variables was different between single and partnered fathers. Single fathers had significantly higher proportion of psychological distress (8.5%) compared to partnered fathers (5.0%). Single fathers were older and were more likely to have an educational level of high school or less compared to partnered fathers. A higher proportion of single fathers were non-regular workers, self-employed, or unemployed than partnered fathers. Single fathers were less likely to have undergone medical examination during the past year and were less likely to drink but more likely to smoke compared to partnered fathers. More single fathers lived in their own home and in a threegeneration or other household compared to partnered fathers. Table 2 shows those with and without psychological distress by demographic characteristics separately for single and partnered fathers. The distribution of father's employment type and sleep hours were significantly different between those with and without psychological distress among single fathers. In the case of partnered fathers, all variables except family type, household expenditure per person per month, age, and number of children had significantly different distribution among those with and without psychological distress. The chi-square test could not be applied to the marital status variable, as all the partnered fathers were married. Table 3 shows the results of logistic regression analysis among single fathers. Employment type and sleep hours were significantly associated with psychological distress. In the crude analysis, unemployed single fathers had significantly higher odds of psychological distress than those with regular employment (OR 3.42; 95% CI, 1.58-7.39). However, the adjusted odds ratio indicated that the association between any unemployment and psychological distress among single fathers was no longer statistically significant (OR 2.19; 95% CI, 0.74-6.50). Single fathers who were selfemployed had significantly lower odds of psychological distress compared to regular employees, even after adjustment for other explanatory variables (OR 0.33; 95% CI, 0.12-0.96). For both crude and adjusted analysis, single fathers who slept less than 6 hours had significantly increased odds of psychological distress (crude OR 2.01; 95% CI, 1.22-3.31 and adjusted OR 2.02; 95% CI, 1.08-3.79) compared to those who slept 6 hours or more. The association between other factors and psychological distress among single fathers was not statistically significant.

RESULTS
Similarly, eTable 2 shows that single fathers who were unemployment, had a history of OPD visit except for depression, had an absence of medical examination during the past year, were non-drinkers, and who were living in a rented home had significantly higher odds of psychological distress than their respective reference groups among partnered fathers. Non-regular workers and unemployed fathers had significantly higher crude odds of psychological distress than regular working partnered fathers. The association for non-regular employment type, however, was not significant after adjustment with covariates. eTable 1 shows the distribution of variables by the child's age category among single fathers. A quarter of fathers of children in the youngest age group were unmarried. A large proportion of fathers with a child aged 12-17 years were self-employed or directors. Fathers of children aged 6-11 years had the lowest proportion (7.8%) of psychological distress compared to fathers of children aged 0-5 years (8.5%) and 12-17 years (8.9%).

DISCUSSION
We found that Japanese single fathers had a higher prevalence of psychological distress compared to partnered fathers. Sleep hours and employment type had a significant association with psychological distress among single fathers. A low proportion of single fathers had a higher educational degree or had undergone medical examination during the past year and were more likely to be unemployed or non-regularly employed or smoke, which was different from partnered fathers.
Our findings showed that the prevalence of psychological distress among single fathers is 8.5%, while that among partnered fathers is 5%. This difference in prevalence is consistent with the findings of Tobias et al 9 in a study of New Zealander fathers and Cooper et al 11 in a United Kingdom study. A study on single mothers found psychological distress among 11% of noncohabiting single mothers. 36 Studies have reported that marriage has a protective effect on men's mental health. [38][39][40] In our study, a large proportion of single fathers were divorced or separated from their spouses. Around 50% of single fathers lived in threegeneration or other households. Amidst a shortage of suitable childcare facilities and lack of flexible working schedules in regular jobs, single fathers live in a three-generation household to provide better care for their children. 41 An additional helping hand might relieve the stress of raising a child alone. However, studies have shown that the stay of single parents in threegeneration households could be short-lived with frequent transitions. 42 Hence, single parents may end up being solely responsible for taking care of their children, as they lack social and mental support. 43 Additionally, having fewer people to trust, depend on, or make connections with may have considerably affected the prevalence of psychological distress among single fathers. 7,9,10,44 The pathway to single parenthood, especially spousal conflict during and after separation, might affect stress levels and the simultaneous coping mechanisms leading to varying mental health outcomes. 45 This study found both increased crude and adjusted odds of psychological distress among unemployed single fathers and decreased odds among self-employed fathers compared to fathers in regular employment. This may be because being unemployed, besides having sole-parenting roles and responsibilities, is a dual burden for single fathers, with an increase in stress. 46 Furthermore, this may be due to the lack of work-life flexibility in full-time regular employment in Japan. The high prevalence of non-regular employment among single fathers is mostly linked to lower-income and social security benefits. 3 With less financial support from the government, single fathers may encounter an increased stress level and low social support for coping.
Our findings showed that sleep of less than 6 hours is associated with increased psychological distress. It is an established fact that sleep plays an important role in mental health balance. 47 Single fathers with young children often juggle lack of sleep and often complain about sleeplessness. 48 Several studies have shown loneliness to be associated with deteriorating health and is also responsible for interrupted sleep. 49 With constant pressure to keep up the financial status of the household, single fathers may have sleep deprivation, which may lead to accumulation of stress. Conversely, psychological distress could have led to poor sleep among these parents. 27 However, we could not establish a causal relationship between sleep and psychological distress because the data we used were cross-sectional.
Our study showed similar findings to the study by Raymo et al, which indicated that single parents tend to have a lower educational level compared to partnered parents. 50 Similarly, a large number of single fathers had non-regular jobs or were unemployed and were more likely to work fewer hours than partnered fathers. Single fathers were also less likely to have undergone a medical examination in the past year and also less likely to consult regarding their stress compared to partnered fathers. 23,24 This suggests a living environment of unstable employment, reduced income, higher stress, and lower levels of Psychological Distress Among Single Fathers in Japan  Dhungel B, et al. health-seeking behavior than partnered fathers, although we did not obtain significant association except for employment. This pattern is particularly concerning given the potentially higher prevalence of key risk factors that we identified.
Of special importance in this context is single fathers' tobacco use. Although we found single fathers are less likely to drink than partnered fathers, we found that 53.6% of single fathers consume cigarettes regularly compared to 40.2% of partnered fathers, which is consistent with the findings of Chiu et al. 15 Prevalence of everyday smoking is high (32%) even among Japanese single mothers living in two-generation households. 36 Studies have confirmed that single parents spend 10-20% of the household income per person on tobacco products. 51,52 The high prevalence of smoking among both single and partnered fathers is concerning and suggests an urgent need to implement programs targeting fathers to reduce nicotine abuse and to further strengthen Japan's tobacco control policies in general. Second-hand smoke exposure could affect their children's health and may be associated with future increased risk of tobacco consumption among these children. 53 A quarter of single fathers of children aged 12-17 years were self-employed or directors. These fathers themselves are more likely to be older than fathers of younger children. Thus, these older fathers are more likely to hold the position of a director following the seniority-based wage system in Japan. 54 Our findings showed that a higher proportion of fathers of children aged 0-5 and 12-17 years had psychological distress than fathers of 6-11-year-olds. As more than half of the fathers of children in the youngest age group were young themselves, they might find parenting stressful. 55 Similarly, as children enter adolescence, they desire autonomy and experience mood swings, which could lead to psychological distress in their fathers. 56 The identified risk factors of psychological distress from this study could be used to identify single fathers at higher risk of psychological distress. Our result shows that single fathers have a higher prevalence of psychological distress compared to partnered fathers, suggesting a possibility of an increase in irregularly employed single fathers with psychological distress. Future policies that aim to reduce mental health problems, including psychological distress, should also focus on singlefather households, as they have numerous existing disadvantages. Strategies and policies need to be formulated to help recognize and mitigate behavioral and lifestyle risk factors among single fathers for both their own wellbeing and that of their children. However, such strategies will be ineffective while single fathers continue to face economic challenges that make healthy lifestyle choices more difficult. In Japan, although educational, financial,  57 suggesting that welfare support is not properly utilized by the small number of single fathers in Japan. There is a need to loosen the eligibility standard for receiving welfare and increase the benefit amount for singleparent families. Free childcare programs have been initiated in Japan since October 2019; however, concerns are being raised over the rapidly increasing number and length of waitlists, as except for low-income families, parents with full-time regular employment are prioritized over parents with non-regular employment. 58,59 Childcare centers should be expanded under high priority with improved day-care environments to provide children with the best care possible. These improved services would help reduce the stress and anxiety of single fathers and ensure that they maximize their efforts caring for their children.
Increased funding and support for the childcare system generally, and expanding the pool of childcare workers, will likely benefit single fathers and should be considered.
To the best of our knowledge, this is the first study that describes the prevalence and risk factors of psychological distress among single fathers and compares it with partnered fathers using a nationally representative data. This study, however, has a few limitations that need to be considered. Although this was a nationally representative study, it only included a small number of single fathers. A study with a larger number of single fathers would help us understand the risk factors better. Additionally, as we utilized cross-sectional data, we could not conclude the direction of causality in the relationship of psychological distress with various employment and behavioral changes, such as sleep hours and smoking habits. Similarly, although this study had a response rate of 77.6%, missing observations might have resulted in selection bias, thus affecting the results of this study. Moreover, the K6 variable that we used as our outcome variable was acquired using a self-administered questionnaire and not through a structured interview. In addition, one should be cautious when generalizing the findings of this study to other countries given the difference in associated cultural values, the structure of the household, and perceived parental roles. Moreover, in Japan after divorce, there is a high probability that a young child gets to stay with the mother. Therefore, we did not have enough data, especially for single fathers with children aged 0-5 years, to do a subgroup analysis by child's age.

Conclusion
In conclusion, the prevalence of psychological distress in single fathers and partnered fathers was 8.5% and 5%, respectively. Single fathers who were self-employed or directors and who slept more than 6 hours were significantly less likely to report psychological distress. As single parents who are self-employed or directors are likely to have significantly reduced psychological distress compared to those with regular jobs, measures are needed to improve the work-family balance for non-self-employed fathers. Additionally, integrated social and health care services should be provided to these fathers to ensure their wellbeing in society. There is a need to provide greater financial assistance and other social welfare support within a framework of a general package of improved welfare rights for single parents to ensure their good health and their children's healthy and fulfilling development.